AWS releases Dogwood Local Engine for agentic governance

AWS has released the Dogwood Local Engine under the Apache 2.0 license to complement its Dogwood governance language. The engine allows developers to embed a library into an enforcement layer that regulates AI agent tool calls based on past actions and outcomes. It specifically supports temporal conditions, such as permitting a git push only if tests passed within a specific recent timeframe.

Temporal governance for AI agents

• Released under the Apache 2.0 license.
• Functions as an embeddable library for an enforcement layer (agent harness).
• Maintains a durable record of past actions, including timing, order, and outcomes to survive system crashes.
• Supports temporal conditions in policies, allowing rules to depend on the history of tool call requests and responses.
• Handles concurrent actions to ensure a consistent view of the event sequence.
• Enables dynamic policy updates without requiring a workflow pause.

Why it matters

This provides an open-source standard for controlling autonomous agent behavior through verifiable, time-sensitive rules.